There are many shortcomings with existing approaches for managing caching technologies, policies and configurations. For example, when workloads share the same controller or disk volume, they also will share the same cache configuration. Unfortunately, this cache configuration often may not be beneficial to all of the workloads.
As another example, with existing approaches caching policies and configuration options can only be changed manually via a management interface. As a result, to adjust cache policies in response to changes in workloads manual intervention is required. Unfortunately, since manual intervention is impractical for frequent small-scale adjustments, resources for caching are typically overprovisioned leading to inefficient use of those resources.